# Lectura de datos #############################################################
URL <- "https://raw.githubusercontent.com/CSSEGISandData/COVID-19/master/csse_covid_19_data/csse_covid_19_time_series/"
URL_confirmados <- paste(URL,"time_series_covid19_confirmed_global.csv", sep = "")
COVID_confirmados_h <- read.csv(URL_confirmados, sep = ",", header = T)
URL_muertes <- paste(URL,"time_series_covid19_deaths_global.csv", sep = "")
COVID_muertes_h <- read.csv(URL_muertes, sep = ",", header = T)
# Preparacion de los datos #####################################################
# Eliminar columnas no usadas
COVID_confirmados_h <- select(COVID_confirmados_h,-c('Lat', 'Long', 'Province.State'))
COVID_muertes_h <- select(COVID_muertes_h ,-c('Lat', 'Long', 'Province.State'))
# cambiar a formato vertical
COVID_confirmados <-
COVID_confirmados_h %>%
gather(fecha,confirmados,2:ncol(COVID_confirmados_h))
COVID_muertes <- COVID_muertes_h %>%
gather(fecha, muertes , 2:ncol(COVID_muertes_h))
colnames(COVID_confirmados) <- c( "pais", "date", "confirmados")
colnames(COVID_muertes) <- c( "pais", "date", "muertes")
COVID_confirmados$date <- as.Date(as.character(COVID_confirmados$date), format = "X%m.%d.%y")
COVID_muertes$date <- as.Date(as.character(COVID_muertes$date) , format = "X%m.%d.%y")
COVID_muertes %>% head
## pais date muertes
## 1 Afghanistan 2020-01-22 0
## 2 Albania 2020-01-22 0
## 3 Algeria 2020-01-22 0
## 4 Andorra 2020-01-22 0
## 5 Angola 2020-01-22 0
## 6 Antigua and Barbuda 2020-01-22 0
# agrupo por pais ( anulo regiones dentro de cada pais)
# otra idea a seguir seria tratar a nivel de region-pais en lugar de solamente pais)
confirmados_por_pais <- COVID_confirmados %>%
group_by(pais, date) %>%
summarise(confirmados = sum(confirmados)
)
muertes_por_pais <- COVID_muertes %>%
group_by(pais, date) %>%
summarise(muertes = sum(muertes))
datos <- merge(confirmados_por_pais, muertes_por_pais)
# genero figura dinamica
#Trazamos las series de tiempo
g1 <- ggplot(datos <- subset(datos, date > "2020-03-15") ,
aes(x = date, y = confirmados, group = pais )) +
geom_line(size = 0.3) +
ggtitle("Confirmados por paĆs") +
scale_x_date(date_breaks = "1 week", date_labels = "%d %b") +
theme(plot.title = element_text(lineheight = 1,face ='bold')) +
ylab("casos confirmados") +
xlab("") +
labs(caption = "\nFuente: The Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
g1 <- ggplotly(g1, tooltip = c("pais")) %>%
layout(legend = list(
orientation = "h",
x = 0.7,
y = 1
)
)
g1
# idem pero con muertes
g2 <- ggplot(datos ,aes(x = date, y = muertes, group = pais )) +
geom_line(size = 0.3)+
ggtitle("COVID_19 - Muertos por paĆs") +
scale_x_date(date_breaks = "1 week", date_labels = "%d %b") +
theme(plot.title = element_text(lineheight = 1,face ='bold')) +
ylab("cantidad de muertos") +
xlab("") +
labs(caption = "\nFuente: The Johns Hopkins University Center for Systems Science and Engineering (JHU CSSE)") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
g2 <- ggplotly(g2, tooltip = c("pais")) %>%
layout(legend = list(
orientation = "h",
x = 0.7,
y = 1
)
)
g2